Mother to child transmission of HIV : maternal and child characteristics.
A prospective study at the University Teaching Hospital of 306 women with their infants, who
were enrolled at delivery, was conducted in 1997. The primary aim was to define the magnitude
and effects of maternal human immuno-deficiency virus (HIV) infection on obstetric problems
and infant outcome.
Women were mainly over 19 years (87.3%), literate (73.7%) and married (91.4%), with no
formal income (75.7%). 48.2% and 46.7% had antenatal or post-partum anaemia (PPA) and of
these 1.8% and 6.2% were severely anaemic. Low post-partum (PP) serum retinol «0.7f.lmoI/L)
and CD4 counts «400 cells/mm3
) occurred in 12.8% and 16.2% of the women. The commonest
obstetric problems were previous child death (32.4%), malaria treatment during pregnancy
(32.6%), previous abortion (16.4%) and hypertension (13.7%).
Post-partum, 30.1% of the women were HIV infected, 14.9% rapid plasma reagin (RPR)
positive and 4.5% hepatitis B surface antigen (HBsAg) positive. Factors independently
associated with HIV infection were: alcohol intake during pregnancy (RR 5.67); ante-partum
haemorrhage (RR 5.85); PP HBsAg positivity (RR 27.45); low PP CD4 cell count (RR 10.63)
and PPA (RR 3.99). Primigravidae had a lower risk ofHIV infection (RR OJ).
For PPA independent risk factors were: caesarean section (RR 9.95); HIV infection (RR 2.81)
and low PP mean corpuscular haemoglobin concentration (MCHC) (RR 8.33); mean
corpuscular volume (MCV) (RR 2.39) and serum retinol (RR 3.03). Alcohol intake during
pregnancy (RR 0.22) and low PP maternal weight (RR 0.10) were associated with reduced risk
The prevalence of low birth weight (LBW; weight <2.5kg), pre-term delivery «37 weeks
gestation) and intra-uterine growth retardation (IUGR; weight < lOth centile for gestational age)
were 18.9%, 23.8% and 25.9%. These showed no association with maternal HIV infection
although the mean birth weight was significantly lower in children born of HIV infected
In HN non-infected women, antenatal anaemia was independently associated with increased
risk pre-term delivery (RR 5.l2) and low birth weight (RR 5.08). Low PP serum retinol
increased the risk of IUGR (RR 3.10).
In HN infected women, lack of paternal income was associated with pre-term delivery (RR
11.7), IUGR with LBW (RR 3.59) and antibiotic treatment in pregnancy with IUGR (RR 5.85).
The cumulative rate of HN mother to child transmission (MTCT) at 1 year of age was 31 %,
with 10.3%, 1O.l% and 9.l% of infants DNA polymerase chain reaction (PCR) positive at birth,
1 month and 4 to 12 months respectively. On multivariate analysis, PP maternal viral load
(>50,OOOcopies Iml) was the only risk factor associated with early infant HN acquisition (birth
and 1 month) (P = 0.005) and cumulative infections at one year (P=O.OOI).
At a year of age, HIV infected children were severely undernourished (weight for age median
Z-score -3.46) and stunted (height for age median Z-score -4.44). Stunting was the main form
of malnutrition in uninfected infants regardless of maternal HN status.
Reported morbidity in infancy was unaffected by HN status. The infant mortality rate was 136
per 1000 live births, 85 per 1000 in HN uninfected children of uninfected mothers, 272 per
1000 in infants of infected mothers and 424 per 1000 in infected infants. After correcting for
confounders, maternal HN infection (HR 0.28) and primigravidae (HR 0.20) were significant
risk factors for infant survival. The population attributable risk percentage of infant mortality
was 41.3% for maternal HN infection and 24.9% when the infant was HN infected as well.